Segmentation Model of Customer Lifetime Value Using K-Means Algorithm

Authors

  • Murnawan

Abstract

This research aims to produce Customer Lifetime Value (CLV) values for each customer segment
using the LRFM method (length, recency, frequency, monetary) and in clustering using the K-Means algorithm.
The clusters produced in this research were 3 clusters. The results of the three segments have been tested for
performance using Euclidean distance. The CLV value will be generated by multiplying the LRFM
normalization value by the LRFM weight value and then adding it up. The sum of the CLV values is carried out
for each cluster that has been formed. The percentage of the number of members in segment 1 is 50% with a
CLV value of 0.3201328, segment 2 is 7% with a CLV value of 0.4646494 and segment 3 is 43% with a CLV
value of 0.2311797. Analysis based on the LRFM value of each segment shows that segment 2 is the segment
that has the best CLV value. This final project produces a visualization of Shiko Outdoor UMKM customer
segmentation with interactive graphics and images in the form of a web-based dashboard.

Published

2020-10-16

Issue

Section

Articles